476 



SCIENCE 



[N. S. Vol. XXXIV. No. 876 



Pteridosperms, are on a higher plane of 

 evolution, and might therefore be expected 

 to show a more highly differentiated type 

 of leaf. But on the contrary these coal- 

 measure plants show a more typically Fili- 

 einean character, both as regards the finely 

 dissected lamina and also in the more deli- 

 cate texture of the foliage compared with 

 the specialized organization of the frond of 

 AcrosticJmm aur-eum, described by Miss 

 Thomas. 



Nor is it necessary to call to aid the 

 salinity of the marsh to explain the excel- 

 lent preservation of the tissues of the plant- 

 remains in the so-called coal-balls, in view 

 of the well-known power of humic com- 

 pounds to retard the decay of vegetable 

 tissues. In addition to these arguments, I 

 might draw attention to the presence of 

 certain fungi among the petrified debris, 

 as more likely to be found in fresh water 

 than in marine conditions. Peronosporites, 

 so common in the decaying Lepidodendroid 

 wood, and the UropJilyctis-like parasite of 

 Stigmarian rootlets, seem to me to support 

 the fresh-water nature of the swamp; just 

 as the occurrence of the mycorhiza, de- 

 scribed by Osborn, in the roots of Cordaites 

 seems to indicate the presence of a peaty 

 substratum for the growth of that plant. 

 Potonie also refers to the occasional occur- 

 rence of Myriapoda and fresh-water shells 

 as indicative of the fresh-water origin of at 

 least many of the coal deposits, and a com- 

 mon feature of the petrified remains of 

 coal-measure plants is the occurrence of the 

 excrements of some wood-boring larva in 

 the passages tunneled by these paleozoic 

 organisms through the wood of various 

 stems. 



A strong argument in favor of the brack- 

 ish nature of these swamps would be sup- 

 plied by the definite identification of Tra- 

 quairia or Sporocarpon as Radiolaria, 

 though we must remember that certain 



marine Coelenterata find their way up into 

 the Norfollc Broads, the fresh-water Me- 

 dusEC are by no means unknown in differ- 

 ent parts of the tropics. Of course, if the 

 coal-measure swamps were estuarine or 

 originated in fresh-water lagoons near the 

 sea, they may have been liable from time 

 to time to invasions of salt water, sufficient 

 to account for the presence of occasional 

 marine animals, but without constituting a 

 halophytic plant association. 



Potonie, who has made so close a study 

 of the formation of coal, and who supports 

 the theory of its fresh-water origin, con- 

 sidered for a long time the comparison 

 between the coal-measure swamp and the 

 cypress swamps of North America, as the 

 nearest but at the same time a somewhat 

 remote analogy, more particularly as he 

 believed that the nature of the coal-meas- 

 ure vegetation required a tropical and also 

 a moister climate than obtains in the south- 

 ern states of North America. Though, in 

 view of the great development of Pterido- 

 phytic vegetation in countries like New 

 Zealand, I think Potonie possibly exagger- 

 ates the temperature factor, he is probably 

 right in assuming a fairly warm climate 

 for the coal-measure forest. The difficulty, 

 so far, has been to account for the great 

 thickness of humic or peaty deposits which 

 must have accumulated for the formation 

 of our coal-seams, in view of the fact that 

 extensive peat formation is generally asso- 

 ciated with a low temperature. In the 

 tropics, peat may be deposited at high alti- 

 tudes, where there is low temperature and 

 high rainfall, but it is generally supposed 

 that the rate of decomposition of vegetable 

 remains is so active that lowland peat- 

 formation was out of the question. Dr. 

 Koorders, however, has observed a peat- 

 producing forest in the extensive plain on 

 the east side of Sumatra, about a hundred 

 miles from the coast. This swamp-forest 



